G3: Genes|Genomes|Genetics Early Online, published on December 23, 2014 as doi:10.1534/g3.114.015149
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Abiotic and biotic stressors causing equivalent mortality induce highly variable
transcriptional responses in the Soybean Aphid
Laramy S. Enders*, Ryan D. Bickel†, Jennifer A. Brisson†, Tiffany M. Heng-Moss*, Blair D.
Siegfried*, Anthony J. Zera§ and Nicholas J. Miller*
*
Department of Entomology, University of Nebraska-Lincoln, Lincoln, NE 68583-0816
§ School
of Biological Sciences, University of Nebraska-Lincoln, Lincoln, NE 68583-0118
† Department
of Biology, University of Rochester, Rochester, NY 14627-0211
1 © The Author(s) 2013. Published by the Genetics Society of America.
17 18 19 20 21 22 23 24 25 26 27 28 Running Title: Aphid transcriptional stress response
Key Words: Transcriptomics, RNA-seq, heat, starvation, plant defense
Corresponding Author: Nicholas J. Miller
Department of Entomology, University of Nebraska-Lincoln,
103 Entomology Hall, Lincoln, NE 68583-0816
[email protected]
Fax: (402) 472-4687
Phone: (402) 472-6200
2 29 30 ABSTRACT
31 Environmental stress affects basic organismal functioning and can cause physiological,
32 developmental and reproductive impairment. However, in many non-model organisms the core
33 molecular stress response remains poorly characterized and the extent to which stress-induced
34 transcriptional changes differ across qualitatively different stress types is largely unexplored. The
35 current study examines the molecular stress response of the soybean aphid (Aphis glycines) using
36 RNA-seq and compares transcriptional responses to multiple stressors (heat, starvation and plant
37 defenses) at a standardized stress level (27% adult mortality). Stress induced transcriptional
38 changes showed remarkable variation, with starvation, heat and plant defensive stress altering the
39 expression of 3,985, 510 and 12 genes respectively. Molecular responses showed little overlap
40 across all three stressors. However, a common transcriptional stress response was identified
41 under heat and starvation, involved with up-regulation of glycogen biosynthesis and molecular
42 chaperones and down-regulation of bacterial endosymbiont cellular and insect cuticular
43 components. Stressor specific responses indicated heat affected expression of heat shock proteins
44 and cuticular components, while starvation altered a diverse set of genes involved in primary
45 metabolism, oxidative reductive processes, nucleosome and histone assembly, and the regulation
46 of DNA repair and replication. Exposure to host plant defenses elicited the weakest response, of
47 which half of the genes were of unknown function. This study highlights the need for
48 standardizing stress levels when comparing across stress types and provides a basis for
49 understanding the role of general versus stressor specific molecular responses in aphids.
3 50 INTRODUCTION
51 Stress is widespread in nature, driving ecological interactions and influencing the
52 evolutionary trajectory of many organisms (Hoffmann and Hercus 2000; Steinberg 2012;
53 Frankham 2005). There are numerous forms of stress, including extreme temperature, drought,
54 pathogens, parasites, and even internal genetic stress caused by the expression of deleterious
55 alleles (Kristensen et al. 2005; Hoffmann and Hercus 2000; Frankham 2005). It is well
56 established that stress can elicit responses across broad categories of biological organization and
57 a wide range of taxa (Kassahn et al. 2009; Bijlsma and Loeschcke 1997; Korsloot et al. 2010).
58 However, because research into stress has progressed independently across several fields of
59 biology, a general framework linking multiple aspects of stress response is currently lacking and
60 limits our understanding of how organisms cope with environmental challenge (Schulte 2014).
61 Approaches are needed that begin to unravel the molecular responses produced by exposure to
62 different forms of stress and that make connections to observed effects on organismal fitness.
63 Organisms are often simultaneously challenged by multiple environmental stresses in
64 their natural environment. The extent to which general versus stressor specific cellular responses
65 exist has therefore been of long-standing interest in ecology and evolutionary biology (Kassahn
66 et al. 2009; Hoffmann and Parsons 1991; López-Maury et al. 2008). It has been hypothesized
67 that organisms have evolved coordinated networks of genes and pathways that respond to a
68 variety of stressors, collectively considered the stress “defensome” (Steinberg 2012), which
69 promote cross-protection and adaption in variable environments (López-Maury et al. 2008;
70 Rangel 2011). The regulation of a core set of genes under a variety of stressors has been
71 demonstrated in bacteria (Battesti et al. 2011), yeast (Gasch et al. 2000; Chen et al. 2003), plants
72 (Ahuja et al. 2010), and animals (Kassahn et al. 2009; Korsloot et al. 2010). Additional research
4 73 supporting the defensome hypothesis indicates a common set of stress responsive proteins exist
74 as well (Tomanek 2011; Wang et al. 2009; Kültz 2005). However, organisms may also require
75 fine-tuned stressor specific responses in order to adapt to changing environments. There is
76 continued debate over the molecular basis of stress specificity, which may be achieved through
77 stress-specific interactions with components of a general defensome, post translational
78 modifications, and/or the compartmentalization of stress proteins (Kültz 2005).
79 Plant-insect systems provide an opportunity to investigate the role of the molecular stress
80 defensome verses stressor-specific responses in promoting adaptation under variable
81 environments. Both partners in these systems suffer direct effects from abiotic stressors such as
82 extreme temperature, but each also serves as a source of biotic stress for the other. The molecular
83 basis of response to stress in plants has been intensively studied (see review by Ahuja et al.
84 (2010)), however insect herbivores are commonly treated as a source of biotic stress for plants
85 rather than active participants in the interaction (Bilgin et al. 2010). As a result, insect stress
86 responses within plant-insect systems are less well characterized, particularly at the molecular
87 level.
88 This study aims to characterize the molecular stress response of the soybean aphid (Aphis
89 glycines). Aphis glycines is a cyclically parthenogenetic species that specializes on soybean
90 (Glycine max) as a host plant. The species is native to east Asia and has become a major
91 agricultural pest in North America since being introduced around 2000 (Hill et al. 2012). Our
92 goal was to measure aphid transcriptional responses to two abiotic stressors (heat and starvation)
93 and host plant imposed biotic stress. Specifically, we investigated the following questions: 1) Do
94 aphids exhibit stress induced transcriptional patterns consistent with the stress defensome
95 hypothesis and 2) To what extent does the magnitude of response elicited at the transcriptional
5 96 level and specific components of the molecular response vary across different stress types? We
97 define stress as causing a significant reduction in organismal fitness relative to benign conditions
98 (Hoffmann and Parsons 1991). We standardized the effect of stress with respect to adult aphid
99 mortality in order to compare molecular responses across qualitatively different stress types.
100 Importantly, because aphids reproduce parthenogenetically, we were able to use a single aphid
101 genotype, such that confounding effects of genetic variation among comparisons between stress
102 types did not influence results.
103 104 105 MATERIALS AND METHODS
Aphid Rearing and Stress Treatments
106 In July 2011 a colony of soybean aphids (A. glycines) was established from a single
107 viviparous parthenogenetic female collected in Madison, Wisconsin. Microsatellite markers
108 developed by Kim et al. (2010) were used to confirm this clonal colony consisted of a single
109 aphid genotype. Aphids were continuously maintained on a single soybean plant (variety
110 KS4202) grown in a plastic Cone-tainer (Ray Leach Cone-tainer, Hummert International, Earth
111 City, MO) and covered by a custom fitted cylindrical plastic cage (30.5 cm x 4.4 cm). Soybean
112 variety KS4202 was used for aphid colony maintenance because it has not been demonstrated to
113 adversely affect aphid survival or development (Enders et al. 2014; Pierson et al. 2010). Soybean
114 plants used for aphid colony maintenance and experiments described below were grown in a
115 greenhouse (16L:8D hour photoperiod). The aphid colony was maintained in a growth chamber
116 at 24±1°C and using a 16L:8D hour photoperiod.
117 Age-synchronized adult aphids were exposed to the following treatments for 36 hours: 1)
118 heat stress (34±1°C) 2) starvation stress 3) plant defensive stress (aphid resistant soybean) and 4)
6 119 benign controls conditions (24±1°C, aphid susceptible soybean). Groups of 20 apterous adult
120 aphids were placed on a single soybean trifoliate (V1 vegetative stage) using a custom built
121 plastic Petri-dish cage (8.9 cm x 2.5 cm) (Enders et al. 2014). For both the control and heat
122 treatments, an aphid tolerant soybean variety was used (KS4202) and for the plant stress
123 treatment, an aphid resistant soybean variety (PI243540) was used that expresses the resistance
124 gene Rag2 (Resistance to Aphis glycines). For the starvation treatment aphids were placed in
125 small Petri-dish cages (35 mm x 10 mm) with a mesh panel in the lid to allow for air circulation.
126 Three experimental blocks were set up, each consisting of twelve replicate groups of 20 aphids
127 per treatment. After 36 hours of exposure to the above four treatments the total number of
128 surviving adults and offspring produced were recorded. Surviving adults were then flash frozen
129 and stored at -80°C. Aphid material harvested from Block I of the experiment was used for
130 transcriptomic analysis (RNAseq) and Block III for reverse transcription quantitative PCR (RT-
131 qPCR) validation of RNA-seq expression levels.
132 Stress intensity or stress level can be quantified by measuring the relative reduction in a
133 measure of interest (e.g. fitness) under stressful and benign conditions (i.e. 1-(Stress/Benign)),
134 such that zero would be no stress and a score of 1 would be the maximum amount of stress (Fox
135 and Reed 2011; Enders et al. 2014). We used aphid mortality to measure stress level. Plant
136 defensive stress caused by Rag2 could not be easily modulated, therefore to achieve a
137 standardized stress level we adjusted the length of exposure to heat and starvation. Preliminary
138 experiments determined all three stressors caused an equivalent increase in mortality relative to
139 control conditions at 36 hours. Our preliminary data and previous work demonstrating cessation
140 of reproduction in starved aphids (Kouamé and Mackauer 1992; Ward and Dixon 1982) and the
141 temperature dependence of insect developmental rates (Ratte 1985) suggests aphid reproduction
7 142 is highly variable under different stresses. Nymph production was therefore considered an
143 unreliable measure of stress and mortality was determined a better indicator. Survival and nymph
144 production were analyzed by ANOVA using the following fixed effects model: Environment
145 (Control, Heat, Starvation, Plant Defense), Block (I, II, III) and Environment x Block. Post hoc
146 multiple comparisons across the four environments were performed using Tukey HSD tests and
147 p values adjusted for multiple comparisons. Under starvation stress aphids did not produce
148 offspring in the 36 hour period of measurement and this treatment was not included in the
149 analysis of nymph production. Raw fitness data is available in Table S1. 150 151 Transcriptomic Methods and Analysis
152 Total RNA was isolated and purified from groups of 32 whole adult aphids using the
153 Qiagen® RNeasy extraction kit according to manufacturer protocols. Three RNA samples were
154 prepared for each of the 4 experimental treatments (3 stresses and control) by randomly pooling
155 aphids from across the 12 experimental replicates. RNA integrity was confirmed using the
156 Agilent 2100 Bioanalyzer and RNA sequencing was performed on the Illumina® HiSeq 2000
157 platform at the University of Nebraska Medical Center Genomics Core facility. Sequencing
158 resulted in 23.5 million total single-end (50 base pair) reads on average per biological replicate.
159 Adapter sequences and low quality sequences were removed prior to further bioinformatic
160 analysis. All raw reads were deposited in the Sequence Read Archive (SRA) at NCBI under the
161 accession number SRP050997.
162 Gene expression was estimated by mapping reads using Bowtie 1.0 (Langmead et al.
163 2009) to the whole adult A.glycines transcriptome assembled by Liu et al. (2012). On average
164 56% of the total reads mapped to the transcriptome across replicates and treatments. An updated
8 165 annotation of the A.glycines transcriptome was performed using the BLAST2GO platform
166 (Conesa et al. 2005), which involved searching contigs against the GenBank non-redundant
167 database using BLASTx algorithms and implementing Gene Ontology (GO) annotation using the
168 Swiss-Prot database and InterProScan. Analysis of differential gene expression was performed in
169 the R statistical environment (R Core Team 2012) using the program DESeq (Anders and Huber
170 2010) with a false discovery rate (FDR) of 0.10. Stress responsive genes were identified by
171 comparing gene expression in the control treatment to each of the individual stress treatments
172 (i.e. three paired comparisons). Enrichment analysis of GO terms associated with genes
173 identified as differentially expressed under each individual stress relative to control conditions
174 was then conducted using Fisher’s Exact Test and GOSeq (Young et al. 2010) at an FDR of 0.05.
175 For starvation stress, only genes with a 2 fold or greater fold change were included in the GO
176 analysis to reduce the dataset to a manageable number. The complete list of GO terms from the
177 analysis of all differentially expressed can be found in supporting information Table S2. Stress
178 responsive genes common to both heat and starvation were separated into two categories: 1)
179 those up- or down- regulated under both stressors and 2) those regulated in opposing directions
180 (e.g. up-regulated under heat and down-regulated under starvation), and corresponding GO
181 enrichment analyses were performed.
182 Gene expression levels were validated using qRT-PCR with five genes identified as
183 differentially expressed under the various stress conditions: HSP70, acyl-protein thioesterase,
184 cathepsin b-2744, 5’-nucleotidase and a cuticular protein. Primer pairs were designed using
185 Primer3 (Rozen and Skaletsky 1999) and qRT-PCR was conducted using SYBR Green® on the
186 BIO-RAD CFX Connect™ Real-Time System (Table S3). Ribosomal protein S9 (RPS9) was
187 used as a reference gene (Bansal et al. 2012) and normalized relative expression levels were
9 188 calculated following methods developed by Hellemans et al. (2007) using inter-run calibrators.
189 Three technical replicates were run per biological sample for each gene.
190 191 192 RESULTS
Effects of Stress on Aphid Fitness
193 All three stressors caused on average a 27% reduction in adult survival relative to benign
194 control conditions, which was consistent across experimental blocks (Figure 1a). Post hoc tests
195 revealed the three stressors had equivalent survival (p values > 0.20) that was significantly lower
196 than under control conditions (p values < 0.05). These effects did not vary across the
197 experimental blocks (Environment x Block: F6,132 = 0.39, p = 0.88). Stress levels with respect to
198 mortality were therefore considered standardized across stress types and experimental blocks.
199 There were significant differences in aphid reproduction under control, heat and plant defensive
200 stress (Environment: F2,99 = 43.95, p < 0.001) and across experimental blocks (Environment x
201 Block: F4,99 = 43.95, p < 0.001). Aphids exposed to plant defensive stress produced on average
202 55% fewer offspring than those under control and heat stressed conditions (p values < 0.001)
203 across all blocks, while heat stressed and control aphids produced similar numbers of offspring
204 (Figure 1b). Starved aphids did not produce offspring over the 36 hour period.
205 206 Transcriptomic Response to Stress
207 Overall, transcriptional stress responses were highly variable despite all three stressors
208 causing an equivalent decrease in adult mortality. The total number of differentially expressed
209 (DE) genes relative to control conditions differed by several orders of magnitude across the three
210 stress types (Figure 2). Starvation had the strongest effect on gene expression (3985 DE genes),
10 211 heat stress had an intermediate effect (510 DE genes), and plant defensive stress induced changes
212 in only a handful of genes (12 DE genes). Consequently, there was only one stress responsive
213 gene common to all stressors, a down-regulated 5’ nucleotidase (Figure 2).
214 Enrichment analysis of stress responsive genes revealed heat stress induced
215 transcriptional changes involved with general stress response, protein refolding and exoskeletal
216 structure (Table 1). Aphid heat stress response was associated with the up-regulation of heat
217 shock proteins (HSP) and the down-regulation of cuticular proteins. Interestingly, molecular
218 chaperones of both the aphid host and its primary endosymbiont (Buchnera aphidicola) showed
219 increased expression. Response to starvation involved increased biosynthesis of basic energy
220 components, nucleosome assembly, and enzymes involved in oxidative processes (Table 1).
221 Starved aphids showed up-regulation of genes involved with glycogen and carbohydrate
222 biosynthesis, histones, cytochrome P450 enzymes and cysteine proteases. DNA replication genes
223 were down-regulated under starvation, as were genes associated with histone methylation and
224 modification. As with heat stress, there was indication of changes in expression of endosymbiont
225 associated genes under host starvation. Several genes associated with the production of
226 peptidoglycan (murein) involved in bacterial cell wall structure showed increased expression in
227 starved aphids.
228 In contrast to the strong transcriptional responses elicited by the heat and starvation, no
229 significant enrichment of biological pathways was found under plant defensive stress, likely due
230 to the low overall number of differentially expressed genes. Plant defensive stress caused the
231 weakest transcriptional response, with only 12 genes responding relative to control conditions
232 (10 up- and 2 down-regulated), of which only six have a predicted function. Two cuticular
233 proteins, acyl-protein thioesterase, a takeout-like protein, and a fatty acid binding protein were
11 234 up-regulated; while a 5’ nucleosidase and gamma-glutamyltranspetidase were down-regulated in
235 response to plant defensive stress.
236 Although we did not find strong evidence for a core transcriptional defensome across all
237 stressors, there were 242 stress responsive genes common to both heat and starvation stress
238 (Figure 2). Among genes that were regulated in the same direction (188), there was enrichment
239 of Buchnera flagellular proteins, cuticular components, and genes associated with protein folding
240 and glycogen biosynthesis (Table 2). Several cuticle proteins and proteins associated with the
241 aphid primary endosymbiont Buchnera were down-regulated under heat and starvation. Both
242 aphid host and primary endosymbiont HSP70 were up-regulated under abiotic stress, as were
243 several genes involved in the production of glycogen, including glycogen synthase. For the 54
244 genes expressed in opposing directions under the two abiotic stressors there was no significant
245 enrichment of associated biological pathways. However, there were several interesting genes
246 with opposing transcriptional responses. Starvation caused a 4-7 fold decrease in expression of
247 several cytochrome P450 enzymes, while heat increased expression 2-3 fold.
248 Overall, validation of transcriptional levels with qRT-PCR on five stress responsive
249 genes using an independent set of biological samples was consistent with results from RNAseq
250 analysis. Stress induced fold changes in gene expression estimated using RNAseq and qRT-PCR
251 were highly correlated (r = 0.96, R2 = 0.91) (Figure S1).
252 253 DISCUSSION
254 Due to the ubiquitous nature of stress, establishing a clear definition and standards of
255 quantification has been challenging. Ambiguity and confusion surrounding the definition of
256 stress has impeded progress towards understanding how organisms cope with environmental
12 257 challenge from different abiotic and biotic stressors. There is continued debate over the use of a
258 threshold level of intensity to define when an environmental factor is considered stressful
259 (Schulte 2014), and studies often classify a condition as stressful despite minimal, or in some
260 cases no impact on fitness or physiological parameters (Fox and Reed 2011). We chose to
261 standardize stress with respect to survival, an approach few comparative studies have attempted,
262 which enabled us to compare the molecular level effects of different stressors. In the current
263 study, a standardized intensity of stress equal to 27% higher mortality relative to benign
264 conditions resulted in 1-25% of the aphid transcriptome responding (Figures 1 and 2). Not only
265 did the magnitude of response at the molecular level vary tremendously, but stressors induced
266 qualitatively different transcriptional responses with little overlap. Our results highlight the
267 complexity of organismal stress responses and a general need for multi-level approaches to
268 understanding how organisms respond and adapt to variable natural environments.
269 270 Do aphids have a molecular stress defensome?
271 Organisms are known to respond to a variety of abiotic and biotic stressors with the
272 coordinated regulation of a core set of genes and pathways (Battesti et al. 2011; Kültz 2005;
273 López-Maury et al. 2008). In insects a number of studies have compared transcriptional
274 responses to multiple stressors in targeted sets of genes (e.g. Freitak et al. 2012; Chen and Zhang
275 2015; Sinclair et al. 2007) , however few have compared global stress induced changes to the
276 entire transcriptome (David et al. 2010; David et al. 2014; Girardot et al. 2004; Sørensen et al.
277 2007). Using next-generation sequencing of the soybean aphid transcriptome, we found a single
278 gene (uridine 5'-nucleotidase) differentially expressed in response to heat, starvation, and plant
279 defensive stress in the soybean aphid (Figure 2). This lack of support for a core set of
13 280 transcriptional changes under the three chosen stressors was likely driven by the relatively few
281 responsive genes identified under plant defensive stress (Figure 2). However, among the abiotic
282 stressors examined there was indication of a common set of transcriptional changes in the
283 soybean aphid (Figure 2), with 188 genes showing similar responses in heat and starvation
284 stressed aphids. The most abundant genes in this group were HSPs, which were up-regulated an
285 average of 1.5 –7 fold relative to control conditions (Table 2). HSPs are a well established core
286 component of the response to a variety of stressors in many organisms, including insects (Zhao
287 and Jones 2012). In aphids HSPs have also been shown to respond to septic wounding and
288 microbial infection (Altincicek et al. 2008), host plant defenses (Francis et al. 2010), and
289 insecticides (Silva et al. 2012).
290 Also up-regulated under both abiotic stressors were genes involved in glycogen synthesis
291 (Table 2). Mobilization of energy stores is predicted to mitigate the costs associated with
292 mounting cellular defenses and overall increased metabolic activity often observed under stress
293 (Silva et al. 2012). Depletion of energy reserves in the form of glycogen has been demonstrated
294 under temperature, moisture and starvation stress in insects (Watanabe et al. 2002; Storey 1997;
295 Gerami 2013). Increased expression of genes involved with the production of glycogen observed
296 in the current study may therefore reflect compensatory mechanisms responding to the depletion
297 of glycogen. Alternatively, increased expression of glycogen synthesis genes may simply
298 represent abnormal transcriptional patterns that could result from direct damage to DNA and/or
299 polypeptides (Wei and Lee 2002) rather than a compensatory or defensive response to stress
300 induced damage or impaired cellular function.
301 302 Several cuticular proteins (CPs) showed decreased expression in response to both abiotic
stressors (Table 2). The insect cuticle plays an important role in protection against environmental
14 303 stress (Neven 2000; Benoit 2010) and CPs have been shown to accumulate during exposure to
304 various stresses in insects (Benoit 2010), including aphids (Nguyen et al. 2009). In contrast to the
305 down regulation of CPs in response to heat and starvation observed in the present study, recent
306 proteomic work in the potato aphid has shown the accumulation of CPs under thermal stress
307 (Nguyen et al. 2009). However, there is a great diversity of CPs in insects, many of which are
308 uncharacterized and their functions unknown (Willis 2010). CPs may therefore be functioning in
309 a stress responsive role that is unrelated to cuticle formation. The production of certain CPs may
310 also be metabolically costly, therefore down-regulation could be linked to conservation or
311 redirection of energy reserves under stress.
312 313 Stressor-specific molecular responses
314 In addition to the large variation in overall magnitude of transcriptional effects across
315 stress types, unique stressor specific responses were also evident (Table 1). Heat stressed aphids
316 showed up-regulation of genes involved in repair of denatured proteins and down-regulation of
317 exoskeletal components. Heat induced damage to proteins and corresponding cellular repair
318 mechanisms are well characterized in a broad range of insects (Neven 2000; Zhao and Jones
319 2012) and enhancement of the cuticular barrier through up-regulation of exoskeletal proteins has
320 been demonstrated under heat stress in aphids (Nguyen et al. 2009). The current study is in
321 agreement with previous work showing up-regulation of HSPs in aphids under stress (Nguyen et
322 al 2009), however we did not find evidence to support heat induced expression of cuticular
323 proteins. This may be because we used adult aphids that had completed their final molt.
324 Starvation response was associated with increased polysaccharide biosynthesis,
325 expression of histones involved in nucleosome organization, and several cytochrome P450s and
15 326 cysteine proteases (Table 1a). These patterns are in agreement with large scale metabolic
327 changes associated with mobilization and shifts in allocation of energy reserves observed under
328 starvation stress (Harbison et al. 2005; Rion and Kawecki 2007). Starved aphids also showed
329 down-regulation of DNA replication and repair mechanisms, a response that has been
330 documented in a variety of taxa (Kassahn et al. 2009; Battesti et al. 2011). Decreased functioning
331 of DNA repair mechanisms under stress is hypothesized to enhance mutation rates as an adaptive
332 strategy in bacteria and yeast (Foster 2007; Galhardo et al. 2007). Reduced expression of HSP90
333 was also observed under starvation stress (Table 1b), a pattern predicted to generate variation
334 upon which natural selection can act in sub optimal conditions (Jarosz and Lindquist 2010). It is
335 currently unknown whether the molecular mechanisms characterized in other organisms that
336 generate novel genetic variation under stress are also present in aphids.
337 Exposure to soybean plant defenses associated with the Rag2 resistance gene elicited the
338 weakest transcriptional response, of which only 6 of 12 genes were functionally annotated.
339 Exposure to soybeans with the Rag2 gene has been shown to adversely affect the behavior,
340 survival and reproduction of the soybean aphid (Hill et al. 2012; Enders et al. 2014). However,
341 the molecular underpinnings of defense pathways associated with soybean resistance (Rag)
342 genes and aphid responses are in the initial stages of molecular characterization (Hill et al. 2012).
343 Our results suggest Rag2 may have a targeted effect on aphids, potentially associated with the
344 production of specific allelochemicals. Interestingly, a homolog to the circadian clock regulated
345 protein takeout (to) in Drosophila was up-regulated in both starved (8-fold) and Rag2 (2-fold)
346 stressed aphids. In Drosophila, takeout is induced under starvation, suggesting this gene may
347 play a role in regulation of energy metabolism and assessment of food availability (Sarov-Blat et
16 348 al. 2000). Rag2 response to soybean aphid feeding may therefore involve mechanisms aimed at
349 both depriving aphids of nutrients and production of defensive toxins.
350 Although molecular chaperones are abundantly expressed under stress, different HSP
351 families perform distinct functions and can demonstrate stress specific regulation in insects
352 (Chen and Zhang 2015; Feder and Hofmann 1999; Zhao and Jones 2012). In the soybean aphid
353 we found comparable HSP70 expression levels under both abiotic stressors, however HSP90 was
354 only differentially expressed in starved aphids (Tables 1 and 2). Similarly, Freitak et al. (2012)
355 found the magnitude of change in expression of three HSPs in Tribolium castaneum varied
356 considerably under heat and starvation stress. Chen and Zhang (2015) have also found that small
357 HSPs respond differently to heat shock, starvation and oxidative stress in the diamondback moth
358 (Plutella xylostella).
359 360 Aphid Endosymbiont Stress Response
361 There is growing evidence that symbiotic relationships play a prominent role in host
362 adaptation to environmental stress (Feldhaar 2011; Gilbert et al. 2010). Insect endosymbionts
363 have been shown to influence parasite and pathogen resistance, heat tolerance, and even
364 manipulate interactions between insects and their host plants (see reviews in Felhaar, 2011 and
365 Frago et al., 2012). However, host-symbiont dynamics under stress remain poorly understood.
366 Several studies indicate the transcriptome of the primary aphid endosymbiont, Buchnera
367 aphidicola, may be relatively stable under various environmental stressors (Moran and Degnan
368 2006; Baumann et al. 1996; Nguyen et al. 2007). However, recent proteomic work in the potato
369 aphid indicated differences in the stress-responsive accumulation of different isoforms of the B.
370 aphidicola chaperone protein GroEL (Nguyen et al. 2009). In the current study multiple B.
17 371 aphidicola heat shock proteins were up-regulated under heat and starvation stress (Table 1).
372 Recent work by Poliakov et al. (2011) confirms that molecular chaperones such as GroEL and
373 DnaK (HSP70) are dominant features of the Buchnera proteome, further suggesting their central
374 role in the protection and maintenance of an endosymbiotic lifestyle (Fares et al. 2004).
375 Several genes involved in metabolism of peptidoglycan, a component of the bacterial cell
376 wall, were up-regulated in starved aphids (Table 1), which could be a defensive response or
377 indicative of bacterial turnover in stressed endosymbiont populations. Endosymbiont flagellar
378 hook and basal body transcripts were also down-regulated under heat and starvation stress (Table
379 2). Genes involved in bacteria flagellular assembly are often lost or undergo functional changes
380 in endosymbionts as a result of adapting to a nonmotile intracellular lifestyle (Toft and Fares
381 2008). While the exact function of these flagellar genes is unknown in Buchnera, it has been
382 suggested they may play a role in protein transport between bacterium and host (Maezawa et al.
383 2006). Decreased expression of flagellar genes may either reflect impaired functioning of
384 Buchnera under stress or an overall reduction in bacterial titer levels.
385 386 Conclusions and Future Directions
387 Adaptation to complex stressful environments likely involves fine-tuned responses as
388 well as a flexible core defense response to a diversity of adverse conditions (Atkinson and Urwin
389 2012; Mittler 2006; Sørensen et al. 2007). Work in Drosophila has shown that both stressor
390 specific and general molecular responses are present in populations selected for tolerance to a
391 variety of abiotic stressors such as heat, starvation and desiccation (Sørensen et al. 2007).
392 However, the overall magnitude of molecular response may vary across multi-cellular
393 organisms, with some exhibiting weak or no global stress response systems (López-Maury et al.
18 394 2008). Results from the current study, as well as work in Aedes (David et al. 2010) and
395 Drosophila (Girardot et al. 2004; Sørensen et al. 2007), suggest core insect responses to abiotic
396 stressors may be comprised of several hundred genes or less. Variation in a number of factors,
397 including stress level and insect life history stage, likely influence detection of global patterns
398 consistent with the stress defensome hypothesis. Fold changes of stress responsive genes should
399 also be considered in comparative studies. Stress-induced fold changes in the soybean aphid
400 ranged from ~1.5 to greater than 20, however in some cases 100-fold and higher up and down-
401 regulation of genes has been observed under multiple stressors in insects (David et al. 2010;
402 Sinclair et al. 2007).
403 In the current study, stressor specific transcriptional responses were prevalent and there
404 was minimal overlap across the different stressors. However, further research exploring response
405 to additional stressors, both alone and in combination, is needed to fully understand the adaptive
406 role of the core aphid defensome versus stressor specific responses. For example, aspects of the
407 shared transcriptional response to heat and starvation may also be present under other stressors,
408 such as those mediated via the host plant or resulting from exposure to xenobiotics. It is also
409 unknown whether fundamental differences exist between abiotic and biotic stress responses. A
410 distinguishing feature of biotic stress interactions is the potential for co-evolution to occur,
411 particularly in plant-insect systems. Identifying similarities and conflicts in response to abiotic
412 and biotic forms of stress could begin to illuminate the molecular basis of long-term adaptive
413 processes in multi-stress environments.
414 Incorporation of additional aphid genotypes is also needed to further our understanding of
415 the ecological and adaptive relevance of the transcriptional stress responses observed in this
416 study. Significant genotype by stress interactions have been demonstrated in aphids for survival,
19 417 reproduction and behavior (Enders et al. 2014; Cardoza et al. 2006; Lombaert et al. 2009; Ferrari
418 et al. 2001). Important differences at the molecular level could potentially underlie intraspecific
419 variation observed under stress at the fitness level. Research exploring the extent to which
420 transcriptional responses are stable across different aphid genotypes will aid in determining the
421 role of a core molecular defensome versus stressor specific responses.
422 Although stress level was equivalent with respect to adult mortality, we found the impact
423 of the different stressors on offspring production varied significantly (Figure 1). Response to
424 stress involves balancing potentially conflicting demands between growth, reproduction, and
425 long-term survival (Atkinson and Urwin 2012; López-Maury et al. 2008), which could contribute
426 to differences across fitness components. During brief periods of starvation, aphids have been
427 shown to delay reproduction (Kouamé and Mackauer 1992; Xu et al. 2012) and in some cases
428 selective reabsorption of embryos is known to occur (Ward and Dixon 1982; Stadler 1995).
429 When deprived of food, aphids potentially allocate resources to maintenance of basic biological
430 functions first and then to reproduction (Kouamé and Mackauer 1992), which is in line with our
431 results showing a lack of reproduction in starved aphids. In contrast, heat stress did not affect
432 reproductive output, which may be explained by generally faster insect development at higher
433 temperatures (Campbell et al. 1974) offsetting the loss of nymphs due to stress induced
434 mortality. Finally, although the exact defensive mechanisms affecting aphid survival and
435 reproduction on Rag2 soybean are unknown, similar reductions in both fitness measures have
436 been previously reported (Enders et al. 2014).
437 Overall, our results demonstrate that equivalent levels of stress imposed on aphid survival
438 can have profoundly different effects on molecular level responses and components of fitness.
439 One explanation for the observed transcriptional differences across stress types is that equivalent
20 440 mortality may be achieved through different mechanisms for each stressor. Variation in the
441 regulation of gene expression affecting the speed of induction and duration of response could
442 also contribute to transcriptional differences across stress types (de Nadal et al. 2011). For
443 example, stress inflicted by Rag2 soybean defenses may elicit strong but transient transcriptional
444 changes earlier than 36 hours, potentially explaining why few genes were found differentially
445 expressed. Alternatively, impaired ability of aphid sensory systems to detect a particular form of
446 plant defensive stress (e.g. allelochemical) could result in delayed or minimal induction of
447 cellular stress response pathways. Overall, our results suggest the magnitude of stress applied,
448 timing of measurement and variation across fitness components should be considered when
449 interpreting results from comparative multi-level stress studies.
21 450 ACKNOWLEDGEMENTS
451 We thank Ashley Yates, Daniel Cloonan, Emma Erikson, and Rebecca Osborn for their help in
452 conducting experiments, George Graef for providing aphid resistant soybean seeds (PI243540)
453 and two anonymous reviewers for helpful comments. This research was funded by a grant
454 awarded through the Life Sciences Competitive Grants Program at the University of Nebraska-
455 Lincoln.
22 456 457 LITERATURE CITED
Ahuja I., de Vos R.C., Bones A.M., and Hall R.D., 2010 Plant molecular stress responses face 458 459 climate change. Trends Plant Sci. 15: 664-­‐674. Altincicek B., Gross J., and Vilcinskas A., 2008 Wounding mediated gene expression and 460 accelerated viviparous reproduction of the pea aphid Acyrthosiphon pisum. Insect 461 Mol. Biol. 17: 711-­‐716. 462 Anders S., and Huber W., 2010 Differential expression analysis for sequence count data. 463 464 Genome Biology 11: R106. Atkinson N.J., and Urwin P.E., 2012 The interaction of plant biotic and abiotic stresses: from 465 466 genes to the field. J. Exp. Bot. 63: 3523-­‐3543. Bansal R., Mamidala P., Mian M.R., Mittapalli O., and Michel A.P., 2012 Validation of 467 reference genes for gene expression studies in Aphis glycines (Hemiptera: 468 Aphididae). J. Econ. Entomol. 105: 1432-­‐1438. 469 Battesti A., Majdalani N., and Gottesman S., 2011 The RpoS-­‐mediated general stress 470 response in Escherichia coli. Annu. Rev. Microbiol. 65: 189-­‐213. 471 Baumann P., Baumann L., and Clark M.A., 1996 Levels of Buchnera aphidicola chaperonin 472 GroEL during growth of the aphid Schizaphis graminum. Curr. Microbiol. 32: 279-­‐
473 285. 474 Benoit J.B., 2010 Water management by dormant insects: comparisons between 475 dehydration resistance during summer aestivation and winter diapause, pp. 209-­‐
476 229 in Aestivation. Springer. 23 477 Bijlsma R., and Loeschcke V., 1997 Environmental stress, adaptation, and evolution: 478 Birkhauser. 479 Bilgin D.D., Zavala J.A., Zhu J., Clough S.J., Ort D.R. et al., 2010 Biotic stress globally 480 downregulates photosynthesis genes. Plant, Cell Environ. 33: 1597-­‐1613. 481 Campbell A., Frazer B., Gilbert N., Gutierrez A., and Mackauer M., 1974 Temperature 482 483 requirements of some aphids and their parasites. J. Appl. Ecol.: 431-­‐438. Cardoza Y.J., Wang S.F., Reidy-­‐Crofts J., and Edwards O.R., 2006 Phloem alkaloid tolerance 484 allows feeding on resistant Lupinus angustifolius by the aphid Myzus persicae. J. 485 Chem. Ecol. 32: 1965-­‐1976. 486 Chen D., Toone W.M., Mata J., Lyne R., Burns G. et al., 2003 Global transcriptional responses 487 of fission yeast to environmental stress. Molecular Biology of the Cell 14: 214-­‐229. 488 Chen X.e., and Zhang Y., 2015 Identification of multiple small heat-­‐shock protein genes in 489 Plutella xylostella (L.) and their expression profiles in response to abiotic stresses. 490 Cell Stress and Chaperones 20: 23-­‐35. 491 Conesa A., Götz S., García-­‐Gómez J.M., Terol J., Talón M. et al., 2005 Blast2GO: a universal 492 tool for annotation, visualization and analysis in functional genomics research. 493 Bioinformatics 21: 3674-­‐3676. 494 David J.-­‐P., Coissac E., Melodelima C., Poupardin R., Riaz M.A. et al., 2010 Transcriptome 495 response to pollutants and insecticides in the dengue vector Aedes aegypti using 496 next-­‐generation sequencing technology. BMC Genomics 11: 216. 24 497 David J.-­‐P., Faucon F., Chandor-­‐Proust A., Poupardin R., Riaz M.A. et al., 2014 Comparative 498 analysis of response to selection with three insecticides in the dengue mosquito 499 Aedes aegypti using mRNA sequencing. BMC Genomics 15: 174. 500 de Nadal E., Ammerer G., and Posas F., 2011 Controlling gene expression in response to 501 502 stress. Nature Reviews Genetics 12: 833-­‐845. Enders L., Bickel R., Brisson J., Heng-­‐Moss T., Siegfried B. et al., 2014 Soybean Aphid 503 (Hemiptera: Aphididae) Response to Soybean Plant Defense: Stress Levels, 504 Tradeoffs, and Cross-­‐Virulence. Environ. Entomol. 43: 47-­‐57. 505 Fares M.A., Moya A., and Barrio E., 2004 GroEL and the maintenance of bacterial 506 507 endosymbiosis. Trends Genet. 20: 413-­‐416. Feder M.E., and Hofmann G.E., 1999 Heat-­‐shock proteins, molecular chaperones, and the 508 stress response: evolutionary and ecological physiology. Annu. Rev. Physiol. 61: 509 243-­‐282. 510 Feldhaar H., 2011 Bacterial symbionts as mediators of ecologically important traits of 511 512 insect hosts. Ecol. Entomol. 36: 533-­‐543. Ferrari J., Müller C.B., Kraaijeveld A.R., and Godfray H.C.J., 2001 Clonal variation and 513 covariation in aphid resistance to parasitoids and a pathogen. Evolution 55: 1805-­‐
514 1814. 515 Foster P.L., 2007 Stress-­‐induced mutagenesis in bacteria. Crit. Rev. Biochem. Mol. Biol. 42: 516 517 373-­‐397. Fox C.W., and Reed D.H., 2011 Inbreeding depression increases with environmental stress: 518 An experimental study and meta-­‐analysis Evolution 65: 246-­‐258. 25 519 Francis F., Guillonneau F., Leprince P., De Pauw E., Haubruge E. et al., 2010 Tritrophic 520 interactions among Macrosiphum euphorbiae aphids, their host plants and 521 endosymbionts: Investigation by a proteomic approach. J. Insect Physiol. 56: 575-­‐
522 585. 523 Frankham R., 2005 Stress and adaptation in conservation genetics. J. Evol. Biol. 18: 750-­‐
524 525 755. Freitak D., Knorr E., Vogel H., and Vilcinskas A., 2012 Gender-­‐and stressor-­‐specific 526 527 microRNA expression in Tribolium castaneum. Biol. Lett.: rsbl20120273. Galhardo R.S., Hastings P., and Rosenberg S.M., 2007 Mutation as a stress response and the 528 regulation of evolvability. Crit. Rev. Biochem. Mol. Biol. 42: 399-­‐435. 529 Gasch A.P., Spellman P.T., Kao C.M., Carmel-­‐Harel O., Eisen M.B. et al., 2000 Genomic 530 expression programs in the response of yeast cells to environmental changes. 531 Science Signaling 11: 4241. 532 Gerami S., 2013 Different exposure methods to neonicotinoids influenced biochemical 533 characteristics in cotton aphid, Aphis gossypii Glover (Hemiptera: Aphididae). Arch. 534 Phytopathol. Plant Protect. 46: 1622-­‐1631. 535 Gilbert S.F., McDonald E., Boyle N., Buttino N., Gyi L. et al., 2010 Symbiosis as a source of 536 selectable epigenetic variation: taking the heat for the big guy. Philos. Trans. R. Soc. 537 Lond., Ser. B: Biol. Sci. 365: 671-­‐678. 538 Girardot F., Monnier V., and Tricoire H., 2004 Genome wide analysis of common and 539 specific stress responses in adult Drosophila melanogaster. BMC Genomics 5: 74. 26 540 Harbison S.T., Chang S., Kamdar K.P., and Mackay T.F., 2005 Quantitative genomics of 541 542 starvation stress resistance in Drosophila. Genome Biology 6: R36. Hellemans J., Mortier G., De Paepe A., Speleman F., and Vandesompele J., 2007 qBase 543 relative quantification framework and software for management and automated 544 analysis of real-­‐time quantitative PCR data. Genome Biology 8: R19. 545 Hill C., Chirumamilla A., and Hartman G., 2012 Resistance and virulence in the soybean-­‐
546 547 Aphis glycines interaction. Euphytica 186: 635-­‐646. Hoffmann A.A., and Hercus M.J., 2000 Environmental Stress as an Evolutionary Force. 548 549 Bioscience 50: 217-­‐226. Hoffmann A.A., and Parsons P.A., 1991 Evolutionary genetics and environmental stress: 550 551 Oxford University Press Oxford. Jarosz D.F., and Lindquist S., 2010 Hsp90 and environmental stress transform the adaptive 552 553 value of natural genetic variation. Science 330: 1820-­‐1824. Kassahn K.S., Crozier R.H., Pörtner H.O., and Caley M.J., 2009 Animal performance and 554 stress: responses and tolerance limits at different levels of biological organisation. 555 Biological Reviews 84: 277-­‐292. 556 Kim H., Kim M.-­‐Y., Kim K.S., Lee H., HOELMER K. et al., 2010 Isolation and characterization 557 of microsatellite loci from the soybean aphid, Aphis glycines Matsumura (Hemiptera: 558 Aphididae). Mol. Ecol. 10: 1098-­‐1105. 559 Korsloot A., Van Gestel C.A., and Van Straalen N.M., 2010 Environmental stress and cellular 560 response in arthropods: CRC Press. 27 561 Kouamé K., and Mackauer M., 1992 Influence of starvation on development and 562 reproduction in apterous virginoparae of the pea aphid, Acyrthosiphon pisum 563 (Harris)(Homoptera: Aphididae). The Canadian Entomologist 124: 87-­‐95. 564 Kristensen T.N., Sørensen P., Kruhøffer M., Pedersen K.S., and Loeschcke V., 2005 Genome-­‐
565 wide analysis on inbreeding effects on gene expression in Drosophila melanogaster. 566 Genetics 171: 157-­‐167. 567 Kültz D., 2005 Molecular and evolutionary basis of the cellular stress response. Annu. Rev. 568 569 Physiol. 67: 225-­‐257. Langmead B., Trapnell C., Pop M., and Salzberg S.L., 2009 Ultrafast and memory-­‐efficient 570 571 alignment of short DNA sequences to the human genome. Genome Biology 10: R25. Liu S., Chougule N.P., Vijayendran D., and Bonning B.C., 2012 Deep Sequencing of the 572 Transcriptomes of Soybean Aphid and Associated Endosymbionts. PloS One 7: 573 e45161. 574 Lombaert E., Carletto J., Piotte C., Fauvergue X., Lecoq H. et al., 2009 Response of the melon 575 aphid, Aphis gossypii, to host‐plant resistance: evidence for high adaptive potential 576 despite low genetic variability. Entomol. Exp. Appl. 133: 46-­‐56. 577 López-­‐Maury L., Marguerat S., and Bähler J., 2008 Tuning gene expression to changing 578 environments: from rapid responses to evolutionary adaptation. Nature Reviews 579 Genetics 9: 583-­‐593. 580 Maezawa K., Shigenobu S., Taniguchi H., Kubo T., Aizawa S.-­‐i. et al., 2006 Hundreds of 581 flagellar basal bodies cover the cell surface of the endosymbiotic bacterium 582 Buchnera aphidicola sp. strain APS. J. Bacteriol. 188: 6539-­‐6543. 28 583 Mittler R., 2006 Abiotic stress, the field environment and stress combination. Trends Plant 584 585 Sci. 11: 15-­‐19. Moran N.A., and Degnan P.H., 2006 Functional genomics of Buchnera and the ecology of 586 587 aphid hosts. Mol. Ecol. 15: 1251-­‐1261. Neven L.G., 2000 Physiological responses of insects to heat. Postharvest Biol. Technol. 21: 588 589 103-­‐111. Nguyen T.T.A., Michaud D., and Cloutier C., 2007 Proteomic profiling of aphid Macrosiphum 590 euphorbiae responses to host-­‐plant-­‐mediated stress induced by defoliation and 591 water deficit. J. Insect Physiol. 53: 601-­‐611. 592 Nguyen T.T.A., Michaud D., and Cloutier C., 2009 A proteomic analysis of the aphid 593 Macrosiphum euphorbiae under heat and radiation stress. Insect Biochem. Mol. Biol. 594 39: 20-­‐30. 595 Pierson L., Heng-­‐Moss T.M., Hunt T.E., and Reese J., 2010 Categorizing the resistance of 596 soybean genotypes to the soybean aphid (Hemiptera: Aphididae). J. Econ. Entomol. 597 103: 1405-­‐1411. 598 Poliakov A., Russell C.W., Ponnala L., Hoops H.J., Sun Q. et al., 2011 Large-­‐scale label-­‐free 599 quantitative proteomics of the pea aphid-­‐Buchnera symbiosis. Molecular & Cellular 600 Proteomics 10. 601 Rangel D.E., 2011 Stress induced cross-­‐protection against environmental challenges on 602 prokaryotic and eukaryotic microbes. World Journal of Microbiology and 603 Biotechnology 27: 1281-­‐1296. 29 604 Ratte H.T., 1985 Temperature and insect development, pp. 33-­‐66 in Environmental 605 606 physiology and biochemistry of insects. Springer. Rion S., and Kawecki T.J., 2007 Evolutionary biology of starvation resistance: what we have 607 608 learned from Drosophila. J. Evol. Biol. 20: 1655-­‐1664. Rozen S., and Skaletsky H., 1999 Primer3 on the WWW for general users and for biologist 609 programmers, pp. 365-­‐386 in Bioinformatics Methods and Protocols. Springer. 610 Sarov-­‐Blat L., So W.V., Liu L., and Rosbash M., 2000 The Drosophila takeout Gene Is a Novel 611 Molecular Link between Circadian Rhythms and Feeding Behavior. Cell 101: 647-­‐
612 656. 613 Schulte P.M., 2014 What is environmental stress? Insights from fish living in a variable 614 615 environment. The Journal of Experimental Biology 217: 23-­‐34. Silva A.X., Jander G., Samaniego H., Ramsey J.S., and Figueroa C.C., 2012 Insecticide 616 resistance mechanisms in the green peach aphid Myzus persicae (Hemiptera: 617 Aphididae) I: a transcriptomic survey. PloS One 7: e36366. 618 Sinclair B., Gibbs A., and Roberts S., 2007 Gene transcription during exposure to, and 619 recovery from, cold and desiccation stress in Drosophila melanogaster. Insect Mol. 620 Biol. 16: 435-­‐443. 621 Sørensen J.G., Nielsen M., and Loeschcke V., 2007 Gene expression profile analysis of 622 Drosophila melanogaster selected for resistance to environmental stressors. J. Evol. 623 Biol. 20: 1624-­‐1636. 624 Stadler B., 1995 Adaptive allocation of resources and life-­‐history trade-­‐offs in aphids 625 relative to plant quality. Oecologia 102: 246-­‐254. 30 626 Steinberg C.E., 2012 Stress ecology: environmental stress as ecological driving force and key 627 628 player in evolution: Springer. Storey K.B., 1997 Organic solutes in freezing tolerance. Comparative Biochemistry and 629 630 Physiology Part A: Physiology 117: 319-­‐326. Team R.C., 2012 R: A language and environment for statistical computing. R Foundation for 631 Statistical Computing, Vienna, Austria. 632 Toft C., and Fares M.A., 2008 The evolution of the flagellar assembly pathway in 633 endosymbiotic bacterial genomes. Mol. Biol. Evol. 25: 2069-­‐2076. 634 Tomanek L., 2011 Environmental proteomics: changes in the proteome of marine 635 organisms in response to environmental stress, pollutants, infection, symbiosis, and 636 development. Ann. Rev. Marine Sci. 3: 373-­‐399. 637 Wang P., Bouwman F.G., and Mariman E., 2009 Generally detected proteins in comparative 638 proteomics–A matter of cellular stress response? Proteomics 9: 2955-­‐2966. 639 Ward S., and Dixon A., 1982 Selective resorption of aphid embryos and habitat changes 640 641 relative to life-­‐span. The Journal of Animal Ecology: 859-­‐864. Watanabe M., Kikawada T., Minagawa N., Yukuhiro F., and Okuda T., 2002 Mechanism 642 allowing an insect to survive complete dehydration and extreme temperatures. J. 643 Exp. Biol. 205: 2799-­‐2802. 644 Wei Y.-­‐H., and Lee H.-­‐C., 2002 Oxidative stress, mitochondrial DNA mutation, and 645 impairment of antioxidant enzymes in aging. Exp. Biol. Med. 227: 671-­‐682. 31 646 Willis J.H., 2010 Structural cuticular proteins from arthropods: annotation, nomenclature, 647 and sequence characteristics in the genomics era. Insect Biochem. Mol. Biol. 40: 648 189-­‐204. 649 Xu X., He S., and Wu J., 2012 The effect of starvation and subsequent feeding on the 650 reproductive potential of the grain aphid, Sitobion avenae. Entomol. Exp. Appl. 144: 651 294-­‐300. 652 Young M.D., Wakefield M.J., Smyth G.K., and Oshlack A., 2010 Method Gene ontology 653 654 analysis for RNA-­‐seq: accounting for selection bias. Genome Biology 11. Zhao L., and Jones W., 2012 Expression of heat shock protein genes in insect stress 655 656 responses. Invertebrate Survival Journal 9: 93-­‐101. 657 658 32 659 TABLES
660 Table 1: Enrichment of GO Terms (FDR < 0.05) and associated stress responsive genes under heat and
661 starvation in A. glycines. Condensed list of GO categories associated with A) genes up-regulated and B)
662 down-regulated under stress. Gene lists correspond to each separate GO term and in cases where genes
663 were associated with multiple terms these GO terms are grouped. The number of gene duplicates is
664 given in parentheses with fold changes relative to control conditions.
33 665 A) Up-regulated Stress Responsive Pathways and Associated Genes
GO ID
GO Description
Gene Name
Fold
Change
HEAT
Response to Stress
GO:0006950
response to stress (BP)
ATP-dependent protease
GO:0006457
protein folding (BP)
activator of HSP 90 ATPase
HSP 70 (7)
HSP 70 (2)
HSP 40 (DnaJ)
1.7
1.4
1.5 - 1.7
18.0 - 24.2
a
1.5
a
HSP 70 (DnaK) (4)
a
HSP 60 (GroEL) (3)
7.3 - 7.7
1.7 - 1.9
stress-induced-phosphoprotein
1.7
aldehyde mitochondrial-like
1.6
acyl-coa dehydrogenase
1.7
caseinolytic peptidase b protein homolog
1.7
CLN3 battenin
2.7
histone partial
2.0
ras homolog gene member c
2.2
suppressor of g2 allele of skp1 homolog
1.6
metastasis suppressor protein 1 (2)
1.5 - 1.6
2.3 - 2.4
STARVATION
Biosynthesis of Basic Energy Components
GO:0000271
polysaccharide biosynthetic process (BP)
1,4 α-glucan-branching enzyme-like (2)
GO:0016051
carbohydrate biosynthetic process (BP)
6-phosphogluconate decarboxylating-like
2.0
GO:0009250
glucan biosynthetic process (BP)
α-glucan-branching enzyme-like
2.1
GO:0005978
glycogen biosynthetic process (BP)
CAMKK-like
2.5
GO:0033692
cellular polysaccharide biosynthetic process
glycogen synthase (2)
2.1
(BP)
GO:0005976
polysaccharide metabolic process (BP)
glycogenin
2.3
insulin receptor substrate
2.2
phosphoenolpyruvate carboxykinase
2.4
Biosynthesis of Endosymbiont Cell Wall Components
GO:0070882
GO:0000270
cellular cell wall organization or biogenesis (BP) D-alanine--D-alanine ligase
peptidoglycan metabolic process (BP)
a
3.5
a
N-acetylmuramoyl-L-alanine amidase (2)
UDP-N-acetylmuramate-alanine ligase
peptidoglycan glycosyltransferase
a
a
2.4 - 3.5
6.6
6.0
Nucleosome Assembly/Organization
GO:0034728
nucleosome assembly (BP)
histone h3 (5)
2.3 - 3.4
GO:0000786
nucleosome organization (BP)
histone h4
GO:0006334
nucleosome (CC)
histone (3)
2.4
2.1 - 2.6
Oxidation-Reduction/NADP Binding
GO:0055114
oxidation-reduction process (MF)
cytochrome P450 (9)
2.1 - 5.6
GO:0016491
oxidoreductase activity (MF)
fatty acyl-reductase (6)
GO:0050661
NADP binding (MF)
glucose dehydrogenase (3)
2.9 - 7.2
2.1 - 4.9
GO:0003958
NADPH-hemoprotein reductase activity (CC)
laccase 1
3.6
GO:0009337
sulfite reductase complex (NADPH) (CC)
short-chain dehydrogenase reductase
2.1
3-oxoacyl-acp reductase (2)
4.5 - 11.4
NADPH cytochrome P450 reductase (3)
2.8 - 3.0
sulfite reductase β-component (3)
12.9 - 24.8
Other Processes
GO:0008234
cysteine protease activity (MF)
cathepsin b (4)
2.0 - 3.4
cathepsin b precursor
2.7
cathepsin b-2744 (3)
4.8 - 5.7
35 666 a
Indicates genes coded by the aphid primary endosymbiont Buchnera aphidicola
667 668 B) Down-regulated Stress Responsive Pathways and Associated Genes
GO ID
GO Description
Gene Name
Fold
Change
HEAT
Exoskeletal Components
GO:0042302
structural constituent of cuticle (MF)
cuticle protein (2)
1.6 - 2.1
cuticle protein precursor (4)
1.8 - 3. 5
RR1 cuticle protein or precursor (3)
endocuticle structural glycoprotein (2)
1.8
1.6 - 1.8
STARVATION
DNA Replication
DNA polymerase delta catalytic subunit
DNA replication (BP)
GO:0006270
DNA-dependent DNA replication initiation (BP) DNA primase large subunit (2)
2.3 - 2.4
GO:0022616
DNA strand elongation (BP)
2.0 - 3.6
DNA replication licensing factor (5)
DNA topoisomerase 2-like
2.0
flap endonuclease 1
2.8
gag-pol polyprotein
3.2
proliferating cell nuclear antigen (2)
replication protein (2)
ribonucleoside-diphosphate reductase large
chain
2.3
GO:0006260
2.5 - 3.0
2.0 - 2.2
2.9
rrm1 protein
2.3
tick partial (2)
2.1
uncharacterized transposon-derived protein
3.4
36 Histone Methylation and Modification
GO:0031061
negative regulation of histone methylation (BP) DNA topoisomerase 2-like
2.0
GO:0031057
negative regulation of histone modification (BP) DNA (cytosine-5)-methyltransferase 1-like
2.4
GO:0051567
histone H3-K9 methylation (BP)
DNA primase large subunit
2.3
2.3
lysine-specific histone
2.0
lysine-specific histone demethylase 1
Other Processes
GO:0006002
fructose 6-phosphate metabolic process (BP)
6-phosphofructokinase (4)
GO:0000398
mRNA splicing, via spliceosome (BP)
RNA-binding protein cabeza
2.1
dead box ATP-dependent RNA helicase
2.4
heterogeneous nuclear ribonucleoprotein h2
2.3
calreticulin
2.6
GO:0051082
unfolded protein binding (MF)
HSP 40 (DnaJ) homolog (2)
2.0 - 2.1
HSP 90 (4)
2.3 - 3.3
j domain-containing protein cg6693-like
2
t-complex protein 1 subunit gamma-like
2.2
669 4.2 - 8.5
37 670 Table 2: Shared abiotic molecular stress response of A.glycines. GO categories (FDR < 0.05)
671 associated with up- and down-regulated genes under heat and starvation stress. The number of
672 gene duplicates is given in parentheses and fold changes are relative to control conditions.
673 38 Fold Change
GO ID
GO Description
Gene Name
HEAT
STARVATION
1.6 - 1.9
1.7 - 2.1
alpha-glucan-branching enzyme-like
1.5
2.1
acyl-coa dehydrogenase
1.7
1.6
HSP 70 (DnaK) [Buchnera aphidicola]
7.3
1.6
1.5 - 1.7
1.4 - 1.6
1.4
1.4
1.5 - 1.6
1.6 - 1.8
1.6
2.3
1.5
2.7
1.5
3.1
Up-Regulated Pathways and Associated Genes
GO:0005978 glycogen biosynthetic process
GO:0006950 response to stress
glycogen synthase (2)
HSP 70 (4)
activator of HSP 90 ATPase
metastasis suppressor protein 1 (2)
GO:0003939 L-iditol 2-dehydrogenase activity sorbitol dehydrogenase (2)
L-threonine ammonia-lyase
GO:0004794 activity
threonine dehydratase deaminase (2)
map kinase-interacting serine threonine-protein kinase
Down-Regulated Pathways and Associated Genes
GO:0009424 bacterial-type flagellum hook
flagellar hook-associated protein 1 [Buchnera aphidicola]
2.6
3.2
GO:0009296 flagellum assembly
flagellar protein flgJ [Buchnera aphidicola]
2.4
2.6
GO:0001539 ciliary or flagellar motility
flagellar basal body P-ring protein [Buchnera aphidicola]
3.5
3.0
GO:0042302 structural constituent of cuticle
acyl- delta desaturase
2.7
2.5
1.6 - 1.8
1.5
GTP cyclohydrolase I
1.5
1.4
RR1 cuticular protein
1.8
1.4
cuticle protein (2)
A
a
100
90
b
Adult Survival (%)
80
b
b
70
60
50
40
30
20
10
0
Control
B
Nymph Production (avg / group)
100
Heat
a
Plant Defense
Starvation
a
90
80
70
b
b,c
c
b,c
60
50
d
40
30
d
d
20
10
0
1
1 2 3
Control
2 3
Heat
1 2 3
Plant Defense
1 2 3
Starvation
Figure 1: Adult survival and reproduction over 36 h in A.glycines exposed to
control conditions and three stressors (plant defense, starvation, and heat). (A)
Adult survival averaged across the three experimental blocks. (B) Average nymph
production is shown for each separate block due to significant variation across
blocks (1,2,3) and environments. Letters indicate significant differences in survival
and reproductive output (p < 0.05).
674 675 40 Plant Defense
1
3733
9
Starvation
1
1
243
266
Heat
Figure 2: Stress responsive genes in A.glycines that are shared and
unique to three stressors: plant defense, starvation, and heat.
Numbers indicate the combined total of genes significantly up- and
down-regulated relative to control conditions.
676 677 41